Association of anti-cyclic citrullinated peptide antibodies, anti-citrullin antibodies, and IgM and IgA rheumatoid factors with serological parameters of disease activity in rheumatoid arthritis.

We evaluated the association of anti-cyclic citrullinated peptide (CCP) antibody titers with serological markers of disease activity. We also compared three different anti-CCP antibody ELISAs with an anti-citrullin ELISA and the IgM and the IgA rheumatoid factor (RF) in their performance of discriminating between rheumatoid arthritis (RA) and other rheumatic diseases. Sera from 333 consecutive patients of the Rheumaeinheit der Medizinischen Poliklinik Munchen, an outpatient clinic for rheumatic diseases, were collected and tested. Anti-CCP antibodies were assayed with three different commercially available ELISAs. Antifilaggrin antibodies were tested with a commercially available ELISA using in vitro deiminated recombinant rat filaggrin. IgA-RF was analyzed with an ELISA, whereas IgM-RF was measured by latex-enhanced turbidimetry. Rheumatoid arthritis (RA) was diagnosed in 87 patients according to the revised classification criteria of the American College of Rheumatology (ACR), probable RA was diagnosed in 23 patients in an early phase not (yet) fulfilling the ACR criteria, and 223 patients had other rheumatic diseases. Differences in sensitivity and specificity were calculated using McNemar's test. A measure of agreement (kappa statistic) was used to examine whether the tests tended to identify the same patients as positive or negative. Correlations between CCP titers and other tests were analyzed by Spearman nonparametric rank correlation. No significant differences in sensitivity and specificity were found between the tested CCP assays (80.0-80.9% and 97.3-98.1%, respectively). All three CCP tests were slightly but not significantly more sensitive and specific than the anti-citrullin assay (77% and 92%, respectively), comparably sensitive but significantly more specific compared with the IgM-RF (86% and 82%, respectively), and significantly more sensitive but comparably specific compared with the IgA-RF (63% and 94.4%, respectively) in detecting the patients with RA. There was no significant correlation between anti-CCP, anti-citrullin, or IgM-RF or IgA-RF antibody titers and C-reactive protein, erythrocyte sedimentation rate, or white blood cell count. A weak but significant linear correlation was found between anti-CCP titers and IgM-RF titers (r = 0.2, P = 0.03). We could not find a significant difference between the three tested anti-CCP assays and the anti-citrullin test in terms of sensitivity and specificity. Compared with the IgM-RF, all the anti-CCP assays were superior in specificity and comparable in sensitivity. Compared with the IgA-RF, they were more sensitive and comparably specific in the discrimination of patients with RA from other rheumatic diseases. No correlation of any tested autoantibody titer with serological parameters of inflammation was found.